Individuals who have disturbed sleep, short sleep, or long sleep are at increased risk for a variety of long-term negative consequences, including early mortality. Individuals who have experienced marital separation or divorce also demonstrate increased morbidity and all-cause mortality over the remainder of life, and there is growing interest in the possibility that disruptions in important health behaviors play a critical role in driving poor health outcomes. Although these broad-based epidemiological effects are well-replicated, relatively little is known about the psychological or behavioral mechanisms that explain (i.e., mediate) the divorce-health association. Using a longitudinal framework that integrates multiple methodologies, the proposed research addresses this limitation by examining psychological distress, sleep, and social engagement outcomes in a sample of 120 participants (from 18 to 70 years of age, equal numbers of men and women). At monthly assessments across 5 months, participants will complete a psychological assessment battery and one week of sleep diaries. At the first, third, and fifth months, all participants will wear a wrist-watch sized actiwatch for one week that measures activity and can be used to derive objective assessment of when and how well individuals sleep. During that week participants also will wear the Electronically Activated Recorder (EAR), which is a naturalistic observation sampling tool that periodically records snippets of ambient sounds from participants' momentary environments. Sampled sound bites are then coded for aspects of participants' interactions that are expected to play a critical role for adjustment to separation and divorce. The research is guided by the following specific aims: (1) To examine how initial levels of sleep disturbance and EAR-indexed behavioral indicators of social engagement moderate changes in psychological distress following divorce; (2) To investigate potential time-varying mediating processes linking sleep, social engagement, and psychological distress; and, (3) to explore potential dynamic associations between sleep and divorce-related psychological distress over the 5 month study period using recent advances in structural equation modeling. The findings will help illuminate how the common life stressor of divorce translates to health/illness outcomes while also providing important basic knowledge about how sleep, social, and psychological changes operate together over time.